A small update. I have just written two extremely simple postscript files: simple1.ps:-
100 100 moveto 200 200 lineto stroke showpage and simple2.ps:- 350 050 moveto 050 350 lineto stroke showpage If I print them in sequence, the bug shows up:- lpr simple1.ps (prints correctly) lpr simple2.ps (prints some of the PJL commands) Running simple2.ps through ghostscript via gs -sDEVICE=hl1250 -sOutputFile=simple2.pcl -r1200x600 simple2.ps and running strings on simple2.pcl shows the PJL commands some of which get printed on the page instead of the graphics: $ strings simple2.pcl %-12345X@PJL @PJL JOB NAME="Ghost" @PJL SET ECONOMODE=OFF @PJL SET MEDIATYPE=REGULAR <== part of this line is printed @PJL SET SOURCETRAY=AUTO <== all printed with offset on 2nd line @PJL SET RESOLUTION=600 <== printed on 3rd line with > offset @PJL SET RAS1200MODE=ON <== 7 leading characters printed on RHS @PJL ENTER LANGUAGE=PCL &l26A &u1200D &l0o0l0E &l1X *p0x0Y *t600R *r1A *b1027M ..[snip].. *b703W *b703W *b703W *b703W *b703W *b703W *b703W *b703W *b703W *b539W &l1T %-12345X@PJL EOJ NAME="Ghost" %-12345X Of course, simple2.pcl is really a mixture of JPL and PCL. Mind you, presumably the bug is in "simple1.pcl" which has not correctly terminated its job leaving the printer in the wrong state to accept simple2.pcl I next want to try sending the files from ghostscript directly to the printer without involving cups at all. Presumably if the bug still shows up, it has to be a gs bug. I have had little time for this so haven't yet tried to get cups to print to a file again, or discover where systemd has hidden it :-( I have yet to read the pcl manual (at least I have it now). After that I guess that I may be able to parse simple1.pcl to see if the PCL commands are properly terminating. IF they are and the direct ghostscript experiment succeeds, then I guess cups/foomatic/whatever must be at fault. Unfortunately I will be away from this printer for several weeks very soon, so this story may be interrupted by a long delay ... As I have said before, all this used to work without problems, but 1 or more years ago. ael -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
